﻿Helium 
  from 
  the 
  Primary 
  Radio-Elements, 
  527 
  

  

  Tespect. 
  On 
  first 
  running 
  the 
  tube 
  the 
  sodium 
  lines 
  were 
  

   strong 
  and 
  D 
  3 
  scarcely 
  visible, 
  but 
  the 
  character 
  o£ 
  the 
  

   discharge 
  suddenly 
  changed 
  and 
  D 
  3 
  came 
  out 
  more 
  strongly, 
  

   and 
  continued 
  for 
  some 
  little 
  time 
  quite 
  obvious. 
  The 
  expe- 
  

   riments 
  with 
  the 
  old 
  type 
  of 
  tube 
  show 
  the 
  great 
  advantage 
  

   of 
  reducing 
  the 
  volume 
  ; 
  for 
  with 
  these 
  the 
  minimum 
  de- 
  

   tectable 
  quantity 
  of 
  helium 
  was 
  about 
  16 
  tenth-grams. 
  It 
  

   was 
  found 
  in 
  actual 
  practice 
  that 
  even 
  when 
  the 
  absorption 
  

   of 
  the 
  residual 
  gas 
  by 
  the 
  calcium 
  was 
  as 
  perfect 
  as 
  possible, 
  

   there 
  was 
  no 
  danger 
  of 
  the 
  helium 
  being 
  so 
  pure 
  that 
  it 
  would 
  

   not 
  conduct 
  the 
  discharge 
  (compare 
  Mackenzie 
  and 
  Soddy, 
  

   Proc. 
  Roy, 
  Soc. 
  1908, 
  80 
  A, 
  p. 
  92). 
  In 
  order 
  to 
  avoid 
  all 
  

   doubt 
  from 
  the 
  fact 
  that 
  the 
  electrodes 
  of 
  spectrum-tubes 
  

   absorb 
  the 
  rare 
  gases, 
  a 
  new 
  spectrum-tube 
  was 
  always 
  used 
  

   for 
  the 
  actual 
  tests 
  with 
  the 
  radio-elements, 
  and 
  in 
  doubtful 
  

   cases 
  in 
  the 
  calibration 
  tests. 
  

  

  In 
  the 
  following 
  Table 
  (p. 
  528) 
  the 
  calibration 
  experiments 
  

   have 
  been 
  collected. 
  The 
  first 
  column 
  gives 
  the 
  number 
  of 
  the 
  

   test 
  (C 
  = 
  calibration). 
  The 
  second 
  column 
  gives 
  the 
  amount 
  

   of 
  helium 
  added 
  in 
  " 
  tenth-grams 
  " 
  (10~ 
  10 
  gram). 
  The 
  third 
  

   column 
  gives 
  the 
  perfection 
  of 
  the 
  experiment, 
  that 
  is, 
  the 
  

   degree 
  to 
  which 
  the 
  absorption 
  of 
  the 
  residual 
  gas 
  was 
  

   complete. 
  If 
  some 
  running 
  of 
  the 
  tube 
  was 
  necessary 
  to 
  

   clear 
  the 
  spectrum, 
  the 
  test 
  is 
  classed 
  as 
  good, 
  fair, 
  or 
  bad, 
  

   according 
  to 
  the 
  amount 
  of 
  running 
  required, 
  Any 
  expe- 
  

   riments 
  in 
  which 
  the 
  spectrum 
  could 
  not 
  be 
  cleared 
  are 
  not 
  

   included. 
  

  

  With 
  regard 
  to 
  XIV, 
  which 
  was 
  a 
  blank 
  test 
  with 
  air 
  

   purposely 
  introduced, 
  it 
  was 
  found 
  impossible 
  to 
  mistake 
  any 
  

   of 
  the 
  lines 
  in 
  the 
  argon 
  spectrum 
  for 
  helium. 
  

  

  Lastly, 
  tests 
  have 
  been 
  made 
  to 
  see 
  what 
  multiple 
  of 
  the 
  

   minimum 
  quantity 
  of 
  helium 
  detectable 
  by 
  means 
  of 
  the 
  D 
  3 
  

   line 
  is 
  required 
  to 
  show 
  the 
  bright 
  red 
  line 
  6677, 
  and 
  the 
  faint 
  

   red 
  7056. 
  In 
  these 
  experiments 
  the 
  calcium 
  furnace 
  was 
  not 
  

   filled 
  with 
  mercury. 
  The 
  minimum 
  quantity 
  of 
  helium 
  to 
  

   show 
  D 
  3 
  in 
  a 
  small 
  lead-glass 
  spectrum-tube 
  without 
  filling 
  

   with 
  mercury 
  was 
  found 
  to 
  be 
  about 
  

  

  3 
  x 
  10- 
  3 
  c.c. 
  ( 
  = 
  5-4 
  x 
  10- 
  7 
  gram). 
  

  

  As 
  the 
  ratio 
  of 
  the 
  volume 
  of 
  the 
  whole 
  apparatus 
  to 
  that 
  of 
  

   the 
  small 
  spectrum- 
  tube 
  was 
  about 
  1 
  : 
  2500, 
  this 
  is 
  in 
  good 
  

   agreement 
  with 
  the 
  preceding 
  tests. 
  Before 
  any 
  glimpse 
  of 
  

   the 
  red 
  line 
  6677 
  was 
  seen 
  three 
  times 
  this 
  quantity 
  had 
  to 
  be 
  

   added, 
  while 
  with 
  four 
  times 
  the 
  line, 
  though 
  still 
  faint, 
  was 
  

   obvious. 
  To 
  bring 
  out 
  the 
  faint 
  red 
  line 
  7056, 
  at 
  least 
  20 
  

   times 
  the 
  minimum 
  quantity 
  is 
  required 
  ; 
  while 
  with 
  30 
  times 
  

  

  